MLM3_p_max: Permutational max test of trait-environment assocation for...

In CajoterBraak/TraitEnvMLMWA: Weighted Averaging (WA) and Multi-Level Methods (MLM) for Trait-Environment Association in Functional and Community Ecology

Permutational max test of trait-environment assocation for MLM models

Description

MLM3_p_max performs the permutational max test of trait-environment interaction for MLM models, starting from a fitted MLM3 object.

Usage

MLM3_p_max(
  MLM3,
  nrepet = 19,
  Binomial_total = 0,
  test_stat = "Wald",
  how_to_permute = list(sites = how(), species = how()),
  print = 1,
  nAGQ = 0
)

Arguments

MLM3	the fitted MLM3 object, created by glmer (lme4) or glmmTMB.
nrepet	number of bootstraps
Binomial_total	scalar, 0 for count-like data and the binomial total for logit models (1 for presence-absence).
test_stat	choice of test statistic; 'Wald' (default) or 'LRT'.
how_to_permute	a list for two how calls, the first for site permutation, the second for species permutation.
print	integer; print progress every print iterations
nAGQ	integer scalar (default 0), used only for an object created by glmer

Details

The code assumes that the interaction parameter is the last fixed parameter in summary(MLM3). The data used in the max test is extracted using dat4MLM2TE_obj. This generates an object of class TE_obj (see make_obj_for_traitenv). dat4MLM2TE_obj is limitted for use with a single trait and single environmental variable, and so is therefore MLM3_p_max. In the model-based permutation tests, either the trait values or the environmental values in the interaction term T*E of the model are permuted to yield a species- and site-level test, respectively. Main effects for the permuted trait and environmental variable are added to ensure that the interaction after permutation has a corresponding main effect. For further details, see Appendices A4 and A5 of ter Braak (2019).

Value

A named list, among which,

p_values	four p-values: one parametric p-value (Wald test) and three permutational p-values: site-based and species-based and the maximum of these two values
obs	values of the test statistic for sites (first row) and species (second row)
sim.row	values of the test statistic for the nrepet data in which the rows of E are permuted
sim.col	values of the test statistic for the nrepet data in which the rows of T are permuted
nrepet	number of permutations

References

ter Braak (2019) New robust weighted averaging- and model-based methods for assessing trait-environment relationships. Methods in Ecology and Evolution (https://doi.org/10.1111/2041-210X.13278)

See Also

expand4glmm.

Examples

## Not run: 
#use a precomputed MLM3 model for the Revisit data
data("MLM3")
## or compute an MLM3 model from the data
# data("Revisit")
# formula.MLM3 <- y ~ poly(env,2) + poly(trait,2) +
  env : trait  + (1 + env|species) + (1 + trait| site)
# MLM3 <- glmmTMB(formula.MLM3, family = betabinomial,  data=Revisit)
summary(MLM3)
res_perm <- MLM3_p_max(MLM3, test_stat = "Wald", nrepet = nrepet, Binomial_total = 100)
names(res_perm)
round(res_perm$p_values,3)

## End(Not run)

CajoterBraak/TraitEnvMLMWA documentation built on Jan. 25, 2023, 7:36 p.m.